Page 1 of 1

H1-5.0-E-G2 - Force Charge firmware bug?

Posted: Fri Oct 04, 2024 3:45 pm
by JimH
Hi

I've been in constant comms with FoxESS for some weeks now and, am assured that all of my inverter and battery settings are at the very latest versions.

The issue I'm still seeing is this :

PV output is clipped to 0kWh issue when Battery SoC reaches 100% DURING a force charge period (only matters in daytime - when there is PV output of course!)
Battery remains at 100% SoC (as it should) but house load is then taken from the grid and available PV output isn’t used either for house load or export.
This remains so until the mode scheduler changes again back to self use from Forced Charge.

However - If the battery is ALREADY at 100% SoC before a force charge period starts then PV output isn’t clipped to 0kWh but battery is maintained at 100% SoC during the Force Charge period by PV output and/or draw from grid.


If I switch off the force charge period manually in FoxESS App V2 the PV output immediately reverts to normal (I can see this in real time via ModBus/HA) - house load first and then export to grid of any surplus. If I switch it back on the bug/issue is there again - i.e. PV output is clipped to 0kWh again.

Surely I should be able to "set and forget" an afternoon topup - especially at this time of year when PV output is so variable without losing a couple of hours of export if the battery has already charged through earlier self-use on sunny days, or only needs a small topup at the start of a Force Charge period??

My system was only installed in February so I don't have much history data to fall back on but as far as I can recall this issue didn't occur prior to the updated firmware release with Octopus Agile integration...

This is how I believe a Force Charge period should work :

When Force Charge period starts:
Battery charges using any available PV output plus draw from grid until it reaches 100% SoC - house load also maintained from this mix of supply by the inverter
When Battery reaches 100% SoC it's held there there until the Force Charge period ends - and any available PV output is used first for house load and any extra is exported to the grid - if there is insufficient PV output (night time or bad weather days) house load is taken from or topped up from the grid.

Am I correct in thinking this is a bug/unexpected behaviour or have I misinterpreted what should happen please?

I'm on Octopus Agile (hence the night and day topup slots - I haven't found the Agile integration reliable enough yet) and here are my Mode Scheduler settings :
IMG_0777.PNG

Re: H1-5.0-E-G2 - Force Charge firmware bug?

Posted: Sat Oct 05, 2024 9:15 am
by Dave Foster
Hi Jim,

I think the problem with the way it works is that we don't actually know what it's supposed to do as it has no documentation, so we're left with what it does to work out why.

Personally I don't use schedules as I have home assistant and i'd rather not trust the cloud connection for anything if at all possible, but i've done testing on how Fox have implemented Force Charge/Discharge and in all honesty it doesn't surprise me when you say it cuts off PV at MaxSoC when force charging.

The schedules appear to use the remote control functions of the inverter rather than them developing a work mode that is built in - the problem then is that the actual MaxSoC (i.e. the one that has always existed in battery settings) is not respected when Force Charging at all.
The get around is they've added MaxSoC to the schedules, but this only works on latest firmware on some inverters - because it is using the remote control functions which are effectively set and forget i.e. force charge at an active power of 4000 watts for 20 minutes, they are having to implement a stop if MaxSoC is reached which i'm guessing is to set active power to 0 - the downside of that is that it will disable PV generation as well.

I guess it's easy for me to say they should have developed a built in work mode for these settings, legacy a number of other constraints probably stopped that from happening so i'm glad they have something, but it has it's downsides.

Can I ask though why you would use Force Charge during the day, normally you just set the inverter in Self Use work mode which will charge the battery in favour of export and does still allow PV generation - admittedly it doesn't charge from grid but I would have thought the grid tariff during the day was not worth doing that (unless you're on a Cosy type tariff with mid afternoon low rate - sorry I don't use agile but wouldn't have thought the rates were reliably low).

Anyway not sure i've answered your question as much as stating my observations with the way it is implemented - and this is only a personal thought but I think the way it is currently implemented is as good as it gets, until they move to an internal work mode that implements it properly.

Re: H1-5.0-E-G2 - Force Charge firmware bug?

Posted: Sat Oct 05, 2024 10:46 am
by JimH
Thanks Dave - that matches my experiences at least.

I use Force Charge windows as above to load shift, as my PV and battery aren't big enough to meet all of my demand on low PV days.

The afternoon force charge window is there (on Octopus Agile) to ensure that my battery is at 100% SoC before the "premium prices" start at 16:00 and to cover me through the later afternoon and evening. The intention is that this is only needed when self-use before 16:00 hasn't topped up the battery to 100% SoC. Grid rates from 14:30 to 16:00 tend to be the lower for day rates on Agile. It's largely redundant on good PV days and there's no issue if 100% SoC is reached before the Force Charge window starts - only if it is reached during the Force Charge window so I lose any export until 16:00.

Inconsistent behaviour at least - which leads me to believe it's a bug/lazy coding issue.

I have HA set up for monitoring and may just have to bite the bullet for control too......

Re: H1-5.0-E-G2 - Force Charge firmware bug?

Posted: Sat Oct 05, 2024 1:55 pm
by JimH
No idea why I didn't think of it before (too focussed on Mode Scheduler and before that Agile Integration I guess) but as there’s some PV today I switched off Mode Scheduler and set my charge from grid times in the “legacy” settings.

Guess what - they work exactly as expected - no clipping of PV output once battery reaches 100% SoC.

I’d say this demonstrates a clear bug/issue with mode scheduler Forced Charge (at least on my H1 Gen 2 inverter)…..

Re: H1-5.0-E-G2 - Force Charge firmware bug?

Posted: Sat Oct 05, 2024 2:12 pm
by Dave Foster
JimH wrote: Sat Oct 05, 2024 1:55 pm No idea why I didn't think of it before (too focussed on Mode Scheduler and before that Agile Integration I guess) but as there’s some PV today I switched off Mode Scheduler and set my charge from grid times in the “legacy” settings.

Guess what - they work exactly as expected - no clipping of PV output once battery reaches 100% SoC.

I’d say this demonstrates a clear bug/issue with mode scheduler Forced Charge (at least on my H1 Gen 2 inverter)…..
Ah sorry I should have mentioned that, yes the normal charge periods do work fine and respect the maxsoc, it’s just a shame the old and new don’t work together.

Re: H1-5.0-E-G2 - Force Charge firmware bug?

Posted: Sat Oct 19, 2024 3:15 pm
by dave600
Hi,

I've also encountered this issue.

Inverter KH9 Inverter with EBS4800 battery stack (installed Sep 2024 so still a bit new to me)

Fox UK updated the Inverter for me and I've now got new column MaxSoC(%). (Previously I was trying to make it work with just the Minimum SoC(%) column.)

I wanted to use the mode scheduler to better use the Octopus Cosy cheep tariff (UK )(04:00-07:00, 13:00->16:00 and 22:00->24:00, with the expensive time 16:00->19:00)

But also still use the solar if it was a sunny day (i.e. Autumn and Spring).
2024-10-19_15h11_59.png
- 04:00-07:00 - Force to 65% (65% should see me through to 13:00 if the sun does not come out)
- 13:00-14:00 - Force to 50% and let the sun charge more if available
- 14:00->15:00 - Force to 70% and let the sun charge more if available
- 15:00->16:00 - Force to 90% and let the sun charge more if available. (90% should see me through the expensive 16:00 to 19:00 period)
- 22:00->24:00 - Force to 65%

I could see the solar clipping while it was charging. See (1) bottom right. I disabled the mode scheduler, clipping stopped and exporting started! This is a screen shot that should highlight this.
2024-10-19_15h18_30.png

I assumed a gentle ramp up of the battery charge would best get me ready for the expensive period but I might time how long it takes to go from 12% -> 100% and set a mode for that to minimise the period that exporting is stopped.

Fingers cross they fix it soon!

Master : 1.38, Slave : 1.01, Manager : 1.33
(With a large heat pump)